Asymmetric hydrogenation of unfunctionalized trisubstituted olefins

Tamara L. Church , Torben Rasmussen , and Pher G. Andersson. Organometallics 2010 29 (24), ..... Christopher M. Haar, Charlotte L. Stern, and Tobin J...
0 downloads 0 Views 254KB Size
J. Am. Chem. SOC.1993,115, 12569-12570

12569

Communications to the Editor Asymmetric Hydrogenation of Unfunctionalized Trisubstituted Olefins with a Chiral Titanocene Catalyst Richard D. Broene and Stephen L. Buchwald'

Scheme I4

-5 mol

1. 2 n-BuLU 0 "C/ 1 atm H2 2. 2.5 PhSIH3

Department of Chemistry Massachusetts Institute of Technology Cambridge, Massachusetts 02139

83-99% ee X2 = 1,1'-binaphth-2,2-dlolate 0

Received July 30, I993 Trisubstituted olefins with suitably disposed chelating substituents can be hydrogenated or isomerized with high levels of enantioselectivity using a chiral cobalt, rhodium, or ruthenium catalyst.'-3 To our knowledge, the application of these catalysts to the asymmetric reduction of unfunctionalized trisubstituted olefins has not been accomplished with good enantioselecti~ity.29~ Chiral titanocene-catalyzed hydrogenation of geminally-disubstituted unfunctionalized olefins was first reported by Kagan. H e showed that 2-phenyl- 1-butene was hydrogenated with low to moderate enantioselectivity using a menthyl-substituted titanocene catalyst.5 The reaction was subsequently optimized and applied to other 1,l-disubstituted olefins by Vollhardt and Halterman, who utilized titanocene derivatives containing novel annulated cyclopentadienyl ligands.6~~ The extension of this reaction to the more challenging asymmetric reduction of trisubstituted olefins has not, to date, been reported. In light of our recent success in the hydrogenation of ketimines using a C2-symmetric titanocene catalyst,8 we decided to examine the corresponding reduction of trisubstituted olefins. In this communication, we report the first highly enantioselective, catalytic hydrogenations of unfunctionalized trisubstituted olefins. The geometric similarity between ketimines and trisubstituted olefins suggested that the catalyst system which was employed for the enantioselective asymmetric reduction of ketimines would be an appropriate starting point for the reduction of olefins.9-ll This catalyst is generated from the C2-symmetric ethylene bis(tetrahydroindeny1)titanocene- lIl'-binaphthyl-2,2'-diolate,1, first reported by Brintzinger.12 Addition of 1.95 equiv of n-BuLi to a T H F solution of (S,S,S)-1 under H2 at 0 OC (Scheme I) generates the active catalyst, which is stabilized by the addition of 2.5 equiv (relative to 1) of phenylsilane. The catalyst mixture (1) Pfaltz, A. Acc. Chem. Res. 1993, 26, 339-345. (2) (a) Noyori, R. Science 1990,248,1194-1 199. (b) Takaya, H.; Ohta, T.; Noyori, R. In Catalytic Asymmetric Synthesis; Ojima, I., Ed.; VCH: New York, 1993; pp 1-39 and references therein. (3) Ojima, I.; Clos, N.; Bastos, C. Tetrahedron 1989,45,6901-6939 and references therein. (4) Tanaka, M.; Ogata, I. J. Chem. SOC.,Chem. Commun. 1975, 735. These workers hydrogenated 2-phenyl-2-butene to 2-phenylbutane with an ee

-_

nf l A.% ~

( 5 ) Cesarotti, E.; Ugo, R.; Kagan, H. B. Angew. Chem., Int. Ed. Engl. 1979, 18, 779-780. (6) Halterman, R. L.; Vollhardt, K. P. C. Organometallics 1988, 7, 883892. (7) Halterman, R. L.; Vollhardt, K. P. C.; Welker, M. E.; Bliser, D.; Boese, R. J . Am. Chem. SOC.1987, 109, 8105-8107. (8) Willoughby, C. A,; Buchwald, S. L. J . Am. Chem. SOC.1992, 114, 7562-7564. (9) Similar zirconium-based catalysts incorporating this ligand have been used in the reduction of 1.1-disubstituted olefins. Waymouth, R.; Pino, P. J . Am. Chem. SOC.1990, 112, 4911-4914 and ref 10. (10) Grossman, R. B.; Doyle, R. A.; Buchwald, S. L. Organometallics 1991, IO, 1501-1505. (1 1) Asymmetric reductions of 1,l-disubstituted olefins using related lanthanide catalysts havebeen reported: Conticello, V. P.; Brard, L.;Giardello, M. A.; Tsuji, Y.; Sabat, M.; Stern, C. L.;Marks, T. J. J . Am. Chem. SOC. 1992, 114, 2761-2762. (12) Wild, F. R. W. P.; Zsolnai, L.; Huttner, G.; Brintzinger, H. H. J. Organomet. Chem. 1982, 232, 233-247.

X2 + l,l'-binaphth-2,2'-diolate

is then combined with substrate (ca. 20 equiv) under an inert atmosphere in a Parr high-pressure reaction vessel, charged to 2000 psig of Hz, and heated for 9-184 h at 65 OC. The results of the hydrogenation reactions are presented in Table I. As is shown in Table I, the reaction works well for both cyclic and acyclic olefins; products are isolated in good chemical and excellent optical yields. We have found that the time and pressure necessary to effect complete reaction are highly substrate dependent. In general, Z olefins are reduced much more slowly than their E isomers. For instance, ( E ) -1,2-diphenylpropene is completely reduced at 80 psig of Hz and 65 O C in 9 h, while only 3% reduction of the Z isomer is observed after 48 h at 70 "C and 2000 psig of H2. Additionally, entries 4 and 5 show that similar Z olefins may be reduced at very different rates; application of our transition-state model (Scheme 11) to these substrates shows that the transition-state structure leading to product in entry 4 is much less hindered than that for entry 5. Reduction of a 64:36 mixture of ( E ) - and (Z)-2-(4-methoxy)-2-butene(entry 3) gave product which was 31% ee. This is the result expected if the individual geometric isomers are reduced with an approximately equivalent level but opposite sense of enantioselectivity. The effective asymmetric hydrogenation of functionalized olefins of the types exemplified by entries 8 and 9 by the chiral rhodium or ruthenium catalyst systems discussed earlier has not been reported. Using our conditions, however, they react to give products with high ee's and in good chemical yield.13 The phenylsilane added prior to the substrate under the standard reaction conditions seems only to stabilize the catalyst during its manipulation in the drybox: it does not appear to be a hydrogen source under the reaction conditions employed. To support this hypothesis, ( E ) -1,2-diphenylpropene was reduced under 80 psig of D2 (20 equiv of substrate, 2.5 equivof PhSiH3, based on catalyst) to provide 1,2-diphenylpropane which was 98% D2 by GC MS. The deuterium was incorporated exclusively at the methylene and methine positions, as evidenced by 2H and I3C NMR, revealing that no olefin isomerization occurred competitively with hydrogenation under these conditions. A difficulty which we initially faced in this work was how to determine the enantiomeric excesses of the products. For most of the examples studied, we found that the ee's of the products could be determined by HPLC using a chiral stationary phase. The methoxy group on the phenyl substituents in entries 2 , 4 , 5 , and 9 was incorporated to facilitate the separation of the enantiomers by HPLC; it was not required for the reaction to be (13) Using rac-1, X = C1, as the precatalyst and I as the substrate, none of the desired product was observed. We speculate that the titanium alkyl resulting from hydride insertion into the olefin undergoes a B-alkoxide elimination reaction. For related processes, see: Buchwald, S.L.; Nielsen, R. B.; Dewan, J. C. Organometallics 1988, 7, 2324-2328.

0002-7863/93/1515-12569$04.00/00 1993 American Chemical Society

12570 J. Am. Chem. SOC.,Vol. 115, No. 26, 1993

Communications to the Editor

Table I' Entry

Starting Materlal

ee

Product

@ @ 1

Me

Yleld (x) 94 91

>99

14'

80

95'

48

79

31

146

80

92

44

77

93

132g

70

83h

184

70

83

169'

Me

&Me

2

>99d >99'

TI^ (h) 48' 9

(W

dMe

Me0 3

Me0 4b

M

64/36 €I2

e

O

w

M

e

&

5

M e O w M e

Meo&

Me0

Ye

Ye 6

a Ph

7

Me0

8

+NBn2

Ih

M

e

O

+NBh

95

43

75

48

80 86

moMe P

9

Me0

O

M

87%

Conversion

m

e

Me0

a Reactions were done at 2000 f 100 psig H2 in THF with 5 f 1 mol % (S,S,S)-l except as noted. Numbers given for ee, time, and yield represent averaged values for two or more runs. Starting material contained 6% 7-methoxy- 1,4-dihydro-2-methylnaphthalene.(R,R,R)-1 was used in this experiment; S (+) isomer was obtained. R (-) isomer wasobtained. 6.7mol% 1. f80psig. g 1800psig. S(+)isomer. 2400 psig, 9.6 mol % 1.

Scheme I1

successful. When it was not possible to employ the chiral HPLC method for the determination of optical purity, the ee's were determined by polarimetric methods.14 We believe that these olefins are reduced through a mechanism similar to that suggested for the reduction of ketimines,8 involving

transition states A and C (Scheme 11). This involves approach of the olefin from the "front" of the complex, as was previously suggested for the hydrogenation of 1,l -disubstituted olefins by a related zirconium catalyst system.9 In this view, the lesssubstituted olefin carbon bonds to the titanium center as the hydride is transferred to the tertiary carbon. This arrangement minimizes the steric interaction between the large substituents on the olefin and the cyclohexyl portion of the tetrahydroindenyl ligand. This view is also consistent with the fact that the rate of reduction for Z olefins (via C) is slower than it is for E olefins (via A), reflecting the energetically unfavorable interaction between one of the large olefin substituents and the ligand in the transition state for the Z olefin. This model suggests that the orientation of the group on the monosubstituted end of the olefin should control the stereochemical outcome of the reduction, as is seen in entry 3, because the steric environment encountered by that carbon is much more demanding than that seen at the other olefinic carbon. This model also predicts the correct absolute configuration of the products in entries 1,2, and 6, as confirmed by comparison to known r0tati0ns.l~The absolutestereochemistry obsewed in the reduction of these olefins is also of the same sense as that seen for the reduction of ketimines with this catalyst system.8 In conclusion, we have described the first examples of catalytic asymmetric hydrogenation of unfunctionalized trisubstituted olefins with high enantioselectivity. The reaction proceeds in good chemical and excellent optical yields. Further work to expand the scope of this catalyst system is in progress.

Acknowledgment. Support for this work was generously provided by the National Institutes of Health (GM 46059). R.D.B. is a trainee of the National Cancer Institute (CA 091 12), which we thank. S.L.B. acknowledges additional support as an Alfred P. Sloan Fellow and a Camille & Henry Dreyfus Teacher-Scholar. Supplementary Material Available: Detailed experimental procedures for the asymmetric alkene hydrogenations, as well as for the preparation and spectroscopic characterization of the starting materials ana products listed in Table I ( l 1 pages). This material is contained in many libraries on microfiche, immediately follows this article in the microfilm version of the journal, and can be ordered from the ACS; see any current masthead-page for ordering information. (14)The enantiomers of racemic entry 1, unfortunately, were not resolved acceptably by either chiral HPLC or GC methods. The rotation found for entry 1, [ a ] D = -78.6', c = 2.02in CHCI3, T = 22 OC (average rotation for three runs using (S,S,S)-l), is greater than the literature value of [ a ] D = -76.3O,c = 2.0in CHCl,, reported both by Paquettel" and in the Dictionary of Organic Compounds;14bwe favor this value because these sources provide rotations for both optical isomers, while the others report a value for only one of the two isomers. We have found two other values for material reported to be optically pure, ranging from [ a ] D = -63.5', c = 2.3 CHCI,, T = 25 OC,lk to [ a ] D = +80.7,c = 2.341 in CHCl,, T = 20 OC.'ld (a) Paquette, L. A.; Gilday, J. P.; Ra, C. S. J . Am. Chem. Soc. 1987, 209, 6858-6860. (b) Dictionary of Organic Compounds, 5th ed.; MacMillan: New York, 1982; Vol. 2, p 2334. (c) Barnes, R. A.; Juliano, B. R. J . Am. Chem. SOC.1959, 81,64624466, (d) Watson, M.B.; Youngson, G. W. J. Chem. SOC.C 1968, 258-262. (1 5 ) See supplementary material for details.